Fufang Xueshuantong for Diabetic Kidney Disease: A Systematic Review and Meta-Analysis

Objective The objective of this meta-analysis was to systematically assess the efficacy and safety of patented Chinese medicine Fufang Xueshuantong (FFXST) for the treatment of diabetic kidney disease (DKD). Methods Randomized controlled trials (RCTs) of FFXST for DKD treatment were searched until May 31, 2020, in seven electronic databases: PubMed, Embase, Cochrane Library, CNKI, Wanfang, VIP, and Chinese Biomedical Literature. The Cochrane risk test from the Cochrane Handbook was used as a bias tool to assess the methodological quality, and Review Manager (RevMan) 5.3 was used to analyze the results. Grading of Recommendations Assessment, Development, and Evaluation (GRADE) criteria were used to classify the quality of evidence. Results Thirteen RCTs involving 1,186 patients were included. The meta-analysis revealed that the efficacy of FFXST in treatment of DKD was significantly superior to that of the control treatment (P=0.0006). The urinary albumin excretion rate (P < 0.01), urinary albumin creatinine ratio (P < 0.0001), and microalbumin (P < 0.0001) were lower in the treatment groups than in the control group. There was also a decrease in low-density lipoprotein cholesterol (P < 0.0001), serum triglyceride (P=0.001), and C-reactive protein (P < 0.0001) in the treatment groups compared with those in the control group. No significant difference in hemoglobin A1c level (P=0.76) and systolic blood pressure (P=0.34) was noted between the treatment and control groups. Three studies reported adverse events, including dizziness and intolerance. In the other 10 trials, adverse events were not mentioned. Conclusion FFXST appears to be effective in the treatment of DKD. However, the low methodological quality of the RCTs suggests that larger, better-designed RCTs are required to verify the clinical eﬀectiveness and safety of FFXST.


Introduction
Diabetic kidney disease (DKD) is one of the most important microvascular complications of diabetes, as well as a key cause of end-stage renal disease (ESRD).It also increases the risk of cardiovascular disease and all-cause death in patients with diabetes [1,2].With the incidence of diabetes increasing annually, the number of DKD cases is also increasing.Approximately 20%-40% of patients with diabetes also have DKD [3].e risk factors of DKD include age, disease course, blood pressure, obesity (especially abdominal obesity), blood lipid, uric acid, and environmental pollutants [4].e main clinical manifestations of DKD are proteinuria and (or) impaired glomerular filtration rate (GFR) [3,5].
Because the occurrence and development of DKD is the result of multifactor interactions, the treatment involves targeting hypoglycemia and hypotension as well as the reduction of proteinuria.Previously, renin-angiotensin-aldosterone system (RASS) inhibitors (angiotensin-converting enzyme inhibitors [ACEI] and angiotensin II receptor blockers [ARB] drugs) have had the most clinical evidence and are recommended as first-line drugs for the treatment of DKD; however, the renal protection effect of RASS inhibitors is limited.More recently, sodium-glucose cotransporter 2 inhibitors have also been recommended for the treatment of DKD.Despite this, a large number of patients with DKD progress to ESRD or die from complications of vital organs outside the kidney annually; thus, it is necessary to develop additional treatment methods to counter DKD [6].
In traditional Chinese medicine, DKD belongs to the categories "cloudy urine" and "edema" [7].In recent years, many Chinese herbal extracts and Chinese patent medicines have demonstrated the reduction of proteinuria and the improvement of renal function in the treatment of DKD.Among them, the Chinese patent medicine Fufang Xueshuantong (FFXST) has been widely used in the treatment of DKD and other diabetic microvascular complications in China.FFXST is composed of Notoginseng Radix, Radix Astragali, Salvia Miltiorrhiza, and Scrophularia Ningpoensis.Studies [8][9][10] have shown that FFXST has a protective effect on the kidneys of diabetic rats, can reduce oxidative stress injury, regulate the RASS system, promote podocyte repair, and improve microcirculation and antiplatelet aggregation.
Although several clinical trials have suggested the efficacy of FFXST for DKD, most of the trials have been singlecenter, including small cohorts and highly different treatment schemes.Hence, it is difficult to verify the clinical efficacy of these treatment strategies.erefore, the goal of this meta-analysis was to assess the efficacy and safety of FFXST for the treatment of DKD, providing evidence for clinical practice.

Search Strategy.
We searched seven electronic databases, including PubMed database, Embase database, Cochrane Library, China National Knowledge Infrastructure database, Wanfang Database, Chinese Science Journal Database, and the Chinese Biomedical Database.We retrieved studies from all of these databases published before May 31, 2020.Our search keywords were as follows: "diabetic kidney disease" OR "diabetic nephropathy" AND "fufang xueshuantong" OR "compound xueshuantong" AND "randomized controlled trial," "controlled clinical trial," "random," "randomly," "randomized" OR "control."Furthermore, we manually searched additional relevant publications according to reference lists from the resulting publications.Different search strategies were applied to Chinese and foreign language databases, without restriction on language or publications.

Inclusion Criteria
(1) Types of trials: randomized controlled trials (RCTs) using FFXST monotherapy or combination therapy with western medicine for the treatment of DKD were included.(2) Types of patients: regardless of the type of diabetes mellitus (DM), stage of the DKD (Mogensen staging criteria), age, gender, or race, we recruited patients who were diagnosed with DKD by clearly defined or internationally recognized criteria.(3) Types of interventions: the experimental group was treated with FFXST monotherapy irrespective of the dosage form or combined with conventional western medicine (ACEI/ARB).ere was no limit to interventions in control groups, whether placebo or ACEI/ARB.Additionally, both groups received routine treatment, such as treatment to lower blood pressure, controlling blood glucose, and regulating serum lipids.(4) Types of outcomes: all included studies that reported at least one of the following outcomes: total effective rate or proteinuria indicators.

Quality Assessment.
Two researchers assessed the risk of bias in trials based on the Cochrane Handbook for the methodological quality of the included studies.We applied the RevMan5.3 to assess the following six items: random sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), incomplete outcome data (attrition bias), selective reporting (reporting bias), and other sources of bias such as baseline comparability of subjects and sample size.We also used the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) approach to assess the quality of evidence for each outcome by GRA-DEpro GDT software.
is classifies evidence as high, moderate, low, or very low quality.Discrepancies were resolved by a third party (Qing Ni).
2.6.Data Analysis.RevMan 5.3 software (Cochrane Collaboration, Oxford, UK) was applied for statistical analysis.Dichotomous data were presented as relative risk (RR), and continuous data were included as the mean difference (MD) or standardized mean difference (SMD) and both included a 95% confidence interval (CI).e heterogeneity evaluations were conducted using a Chi 2 test.e fixed-effects model was used when the heterogeneity was significant (P > 0.10, I2 ≤ 50%); otherwise, a randomized effects model was used (i.e., when P < 0.10, I2 ≥ 50%).
e possible sources of heterogeneity were explored by sensitivity analysis and 2 Evidence-Based Complementary and Alternative Medicine subgroup analysis.Publication bias was tested using funnel plots when the number of experiments was ≥10 [11].

2.7.
Outcomes.e primary outcome indicator was a total effective rate, which was based on changes in symptoms and the level of proteinuria [12].e total effective rate was categorized as significantly effective cases (urinary albumin excretion rate [UAER] returned to normal levels or decreased by more than 50%, with an obvious improvement in symptoms), effective cases (UAER decreased by less than 50%, improvement in symptoms), or ineffective cases (no improvement in either UAER and symptoms).

Study Search and Selection.
Initially, a total of 114 publications were identified from the seven electronic databases.After removing 67 duplicate publications, we excluded 20 nonclinical studies by reading the titles and abstracts.After a full-text review, we excluded two studies with significant data errors, three nonrandomized controlled studies, and nine interventions or outcome indicators that did not meet inclusion criteria.Finally, 13 studies were included in this meta-analysis.e retrieval process is shown in Figure 1.

Characteristics of the Included Studies.
Characteristics of the 13 studies [13][14][15][16][17][18][19][20][21][22][23][24][25] are summarized in Table 1.All the included studies were published between 2009 and 2019.e 13 RCTs involved 1186 subjects (592 in treatment groups and 594 in control groups), and the sample size for each study ranged from 48 to 130 subjects.In terms of the disease type and stage, patients with type 2 diabetes (T2DM) were included in eight studies, whereas the remaining five studies did not report in detail the type of diabetes patients included.Except for one study [14] that did not report the DKD stage, the remaining 12 studies included subjects who were DKD patients in Mogensen III according to the Mogensen stage.e subject's DKD diagnosis was clear in 13 studies.All studies included patients who met the diagnostic criteria for DKD, nine of which used the criteria of the World Health Organization (WHO) DM diagnostic criteria [26] and Mogensen diagnostic [27].One study [18] referred to the American Diabetes Association criteria combined with the Epidemiology and Diagnostic Criteria of Diabetic Nephropathy [28].Another [16] used the WHO DM1999 and pathological diagnosis of renal biopsy.e diagnosis of DKD in one study [14] was based on internal diagnostic criteria for diabetes combined with symptoms of proteinuria and history of diabetic retinopathy and the internal DKD diagnostic criteria were used in another study [24].
Compared with the control group, treatment groups in six RCTs [16,17,[22][23][24][25] were treated with FFXST monotherapy, whereas treatment groups in the other seven RCTs received FFXST combined with ACEI/ARB.All patients in both groups were treated with conventional hypoglycemic therapy.e duration of the trials ranged from 8 weeks to 24 weeks.
Only two studies used the total effective rate based on changes in symptoms and urinary protein levels as the main outcome indicators.In terms of proteinuria indicators, five studies reported UAER, six studies reported ACR, and six studies reported mAlb.Additionally, we also used BUN, HbA1c, standard bicarbonate (SBP), TG, LDL-C, and C-reactive protein (CRP) as secondary outcome indicators.Adverse events were not mentioned in three studies [15,17,21], and none of the studies reported a decrease in the quality of life or took adverse indicators (e.g., deterioration rate, access to dialysis rate, etc.) as outcome measures.

Risk of Bias in the Included Studies.
e methodological quality assessment of the 13 RCTs is shown in Figures 2 and  3. Two studies [16,17] adopted methods of randomization using a random number table; one study [20] used mechanical random sampling, and seven studies [13,14,18,19,22,23,25] only mentioned "randomization" but did not describe specific methods.e remaining three RCTs had a high risk of selection bias, because two [15,21] followed the order of medical treatment, and one [24] followed the case number.None of the 13 studies were double-blind, and no study indicated details on allocation concealment or sample size calculations.Two studies [22,23] showed high-risk bias in selective reporting.Baseline information was similar for different groups of subjects in all 13 studies.In short, the quality of all RCTs was generally low and contained a risk of bias.

Total Effective Rate.
e total effective rate was reported in two studies, and the results indicated significant differences between the two groups.ese trials exhibited nonsignificant heterogeneity (χ 2 � 0.34, P � 0.56, I2 � 0%); thus, the fixed-effects model was used for statistical analysis.

Evidence-Based Complementary and Alternative Medicine
To further compare the differences in UAER between FFXST combined with conventional western medicine and control groups, subgroup analysis was performed.In one study [17], FFXST monotherapy in the treatment group was compared with that of the control group (MD � −51.60, 95% CI: −64.04-39.16,Z � 8.13, P < 0.00001).FFXST combined with ACEI/ARB was compared with control groups in the remaining four studies (N � 305, MD � −24.61, 95% CI: −40.21-9.02,Z � 3.09, P � 0.002), without significant heterogeneity (χ2 � 13.44, P � 0.004, I2 � 78%).Sensitivity analysis suggested that the heterogeneity between the subgroups decreased significantly after removal of one study [21] (χ2 � 0.66, P � 0.72, I2 � 0%).e test method and kit used in this study for the detection of UAER indicators may have been a source of heterogeneity.

Glycemic Control.
In this review, we mainly assessed HbA1c.

Blood Lipid.
is meta-analysis evaluated two blood lipid indicators, LDL-C and TG.Four studies reported that the LDL-C indictor had low heterogeneity (χ 2 � 3.02, P � 0.39, I2 � 1%), and a fixed-effect model was adopted.It was determined that the LDL-C level of treatment groups was lower than that of the control groups (MD � −0.39, 95% CI: −0.58-0.20,Z � 4.05, P < 0.0001) (Figure 10).
e results indicated there was no statistical difference in TG improvement between the two groups (N � 613, MD � −0.36, 95% CI: −0.59-0.13,Z � 3.11, P � 0.002) with heterogeneity (χ 2 � 24.17, P � 0.0005, I2 � 75%).Sensitivity analysis showed that heterogeneity decreased after the removal of two studies (Dai XM, 2012, and Yun P, 2013) (χ 2 � 4.05, P � 0.40, I2 � 1%).However, the TCM syndrome in the subjects in the two studies was the syndromes of Qi and Yin deficiency combined with stasis, and the remaining studies were not limited to TCM syndromes.us, TCM syndromes may have been the source of heterogeneity.

Adverse Events.
ree of the 13 RCTs mentioned adverse events in treatment groups.No adverse reactions occurred in the treatment group one study (Lu YZ, 2009) Evidence-Based Complementary and Alternative Medicine the treatment group of two studies (Wang ML, 2017, and Yun P, 2013), five patients had dizziness and three patients had a poor appetite.Adverse events were not reported in the other 10 RCTs; therefore, the safety of the FFXST therapy needs further evaluation.
3.4.11.Publication Bias.We could not conduct the funnel plot analysis for the detection of publication bias because of an insufficient number of experiments.

Grade Assessment.
According to the GRADE, the quality of evidence was rated as moderate for the primary outcome, and the secondary outcomes were rated as low, very low, or moderate (Table 2).

Discussion
DKD is one of the common microvascular complications of DM.Early prevention and treatment can delay the occurrence and progression of DKD, which is of great significance 407 Total (95% Cl) 404 100.0 -0.39 [-0.63, -0.15] Heterogeneity: tau 2 = 0.10; chi 2 = 51.42,df = 8(P < 0.00001); I 2 = 84% Test for over all effect: Z = 3.21 (P = 0.001) Test for subgroub differences: chi 2 = 0.00, df = 1 (P = 0.99), I to the improvement, survival rate, and quality of life of diabetic patients [29].Clinical practice shows that TCM has the characteristics of multitarget, multipathway, and low adverse reaction [30][31][32] and has a great potential in the intervention of DKD.FFXST is a Chinese patent medicine approved by the State Food and Drug Administration of China and has been widely used for the treatment of DKD. is meta-analysis suggests that FFXST may be a safe and effective treatment for DKD.
is meta-analysis revealed that FFXST for DKD was superior to the treatments provided the control group in total effective rate.Additionally, FFXST exhibited advantages in improving proteinuria indicators (UAER, ACR, and mAlb), blood lipid (LDL-C, TG), and inflammatory index (CRP), but not in lowering BUN or HbA1c levels or blood pressure in DKD patients.However, it has been reported that FFXST has a certain antihypertensive effect, which needs to be confirmed by further studies in the future [33].
ere were many limitations to our meta-analysis.Although the dosage and form of FFXST used in the 13 studies were consistent, the treatment periods, DM type, and DKD stage of patients were not similar among the RCTs.Furthermore, the methodological quality of the studies was generally low, and the sample size was not reported in the 13 RCTs.All of the previously mentioned factors might have negatively affected the reliability of the research results.
Regarding the outcome, only two in the 13 RCTs reported the total effective rate, which was the primary outcome in our review.e improvement of the symptoms of patients is an important part of the evaluation of the efficacy of DKD treatment, but only one study (Lu YZ, 2009) reported the change of symptom score of patients before and after treatment.us, we could not evaluate the improvement effect of FFXST on the symptoms of patients.It was necessary to standardize the DKD efficacy evaluation system in clinical trials, which could have improved the reliability of the analysis.One study (Dai XM, 2012) also showed that FFXST could improve hemorheology.However, such reports were rare, and more pharmacological and clinical studies are needed to verify the mechanism of FFXST in the treatment of DKD.Clinical events are often recommended as primary outcome indicators for clinical studies; however, no trials assessed the incidence of DKD clinical endpoint events (death/entry to ESRD) or other adverse indicators in our study, which may not be conducive to explain the effect of FFXST for DKD.
e GRADE results showed that the evidence quality of the total effective rate and LDL-C level was moderate, and the quality of the remaining outcomes was low or very low.
is is mainly due to the fact that most of the included studies did not use blind methods or the large heterogeneity between studies.erefore, more rigorous clinical studies are still needed to confirm the efficacy of FFXST in the treatment of DKD.
Follow-up and adverse event reports were insufficient among the 13 studies, with only three studies reporting adverse events and one study reporting a follow-up; thus, this meta-analysis was unable to assess the long-term efficacy and safety of FFXST for DKD.

Conclusions
Our meta-analysis suggested that the Chinese patent medicine FFXST was superior to that of the treatment of the control group in the improvement of total effective rate, reduction of proteinuria, and lowered blood lipid.DKD patients, especially who are in the stage of Mogensen III, accompanied by abnormalities in indicators of UAER, ACR, mAlb, LDL-C, TG, and CRP, can be treated with FFXST or combined with western medicine.However, FFXST may not be an optimizing option to improve abnormal indicators of SBP, HbA1c, and BUN in DKD patients.
However, the long-term efficacy and safety of FFXST for DKD is uncertain because most studies included in this review were of low quality, having small sample sizes and high heterogeneity.
us, high-quality, large-scale, and multicenter RCTs are needed to validate the current results.

Figure 1 :
Figure 1: Flowchart of the literature retrieval.

8
Two researchers extracted the information independently.e data included study ID, baseline patients, disease data, interventions, and outcomes (e.g., sample size, age, gender, type of DM, stage of DKD, interventional measures, treatment duration, reporting of adverse events, and outcome measures).Discrepancies were resolved by discussion with other authors.

Table 1 :
Characteristics of the included studies.-Based Complementary and Alternative Medicine the subgroup sensitivity decreased (χ 2 � 14.06, P � 0.032, I2 � 12%).e sources of heterogeneity may have been related to the conventional treatment regimen adopted in the study, where ACEI/ARB drugs were preferred for patients with hypertension, leading to some patients belonging to the FFXST plus ACEI/ARB subgroup, rather than FFXST subgroup alone.
. In Figure 11: Forest plots of FFXST effects on TG.