Efficacy and safety of Huangkui capsule for diabetic nephropathy: A systematic review and meta-analysis

Abstract

Background:

To systematically evaluate the efficacy and safety of Huangkui capsule (HKC) in the treatment of diabetic nephropathy (DN).

Methods:

A computerized search of 8 online databases, CNKI, VIP, WANGFANG, CBM, ChiCTR, Embase, PubMed, and Cochrane Library, was conducted from the time of database construction to June 23, 2023, and randomized controlled trials of HKC for the treatment of DN were included according to the predefined screening criteria. Literature quality was evaluated with the help of Cochrane Risk of Bias Assessment Tool, and data were analyzed by STATA/MP17 and RevMan 5.3 software.

Results:

A total of 1543 articles were retrieved, and 45 studies were finally included, all in Chinese, with a total sample size of 4297 cases, including 2168 cases in the experimental group and 2129 cases in the control group, and the quality of all the included studies was generally low. Meta-analysis showed that (1) efficacy: effective rate: [relative risk = 1.24, 95% confidence interval (CI) (1.20, 1.29), P < .001], SCR: [standard mean difference (SMD) = −1.19, 95% CI (−1.47, −0.91), P < .001], 24 h-UTP: [SMD = −1.27, 95% CI (−1.66, −0.88), P < .001]. Other renal function, glucose metabolism, lipid metabolism, and inflammatory factors related indicators improved compared with the control group. But in the outcome index of estimated glomerular filtration rate, the addition of HKC did not have an advantage over conventional treatment alone. Estimated glomerular filtration rate: [SMD = −0.21, 95% CI (−0.80, 0.39), P = .50]. (2) Safety: There was no statistical significance in the incidence of adverse reactions between the study group with the addition of HKC and the control group.

Conclusion:

The clinical efficacy and safety of DN treated with the combination of HKC is better than that of conventional western medicine alone, but to better evaluate the efficacy and safety of HKC as an adjunctive intervention for DN, more rigorously designed large-sample, multicenter randomized controlled trials are needed to provide evidence support in the future.

1. Introduction

Diabetic nephropathy (DN) is the most common microvascular complication of diabetes mellitus (DM), the most common cause of end stage renal disease, and an important cause of disability and death in DM.^[1] DN clinical manifestations mainly include increased renal artery pressure, glomerular hyperplasia, basement membrane thickening, glomerular hyperperfusion, and renal dysfunction.^[2] Four hundred sixty-three million people worldwide have DM in 2019, expected to rise to 578 million in 2030 and 700 million in 2045.^[3] In addition, 20% to 40% of patients with DM have a combined DN.^[4] Therefore, the prevention and treatment of DN has become a public health concern. At present, western medicine is the mainstay of DN diagnosis and treatment, and it is mainly used to slow down the course of the disease by changing bad lifestyle, controlling blood sugar, lowering blood pressure, regulating lipids, and improving microcirculation. However, the treatment often fails to achieve satisfactory results, and long-term use of medication is prone to cause adverse reactions, such as allergies and liver and kidney damage.^[5] For this reason, researchers at home and abroad have begun to explore the interventional treatment of DN with traditional Chinese medicine. Studies have shown that traditional Chinese medicine can significantly improve the quality of life and clinical indicators of DN patients, slow down the progression of DN, and have fewer adverse drug reactions.^[6,7]

In the actual clinical application and research, it is found that Huangkui capsule (HKC) has good efficacy in the treatment of DN, which can delay the progression of renal function, improve proteinuria, and reducing inflammatory damage, etc. However, most of the published clinical studies of HKC for the treatment of DN are single-center studies, with small sample size, bias, and insufficient description of adverse effects, providing a low grade of evidence-based medicine. Based on evidence-based medicine, the present study conducted meta-analysis of domestic and international clinical randomized controlled trials (RCTs) on the treatment of DN with HKC to clarify the clinical efficacy and safety of HKC in the treatment of DN, and to provide evidence-based medical evidence for clinical application. Therefore, the aim of this study is to systematically evaluate the efficacy and safety of HKC in the treatment of DN, in order to provide more reliable evidence for the clinical application of HKC.

2. Methods

This systematic review followed the methods of the Cochrane Handbook for Systematic Reviews of Interventions (version 6.3) and complied with the 2009 Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement guidelines.^[8,9] We registered the review protocol with PROSPERO at the beginning (CRD42023439355).

2.1. Search strategy

The 2 researchers searched the following Chinese and English databases: China National Knowledge Internet (CNKI), VIP Information Chinese Periodical Service Platform (VIP), Chinese Biomedicine Literature Database (CBM), Wanfang Data Knowledge Service Platform (WANGFANG), China Clinical Trials Registry Center (ChiCTR), The PubMed database, Embase database and the Cochrane Library were searched from the date of creation of the databases to June 23, 2023. The Chinese database is searched by a combination of subject terms and titles/keywords, and the search terms include “huang kui jiaonang,” “Huangkui,” “huangshukuihua,” “diabetic nephropathies,” “diabetic kidney disease,” “diabetic nephropathy.” The English database was searched by combining subject terms with free words. Search terms include “huangkui capsule,” “huangkui jiaonang,” “huangkui,” “Abelmoschus manihot,” “Diabetic Nephropathies,” “Diabetic Nephropathy,” “Diabetic Kidney Disease,” “Diabetic Kidney Diseases,” “Kidney Disease, Diabetic,” “Kidney Diseases, Diabetic,” “Nephropathies, Diabetic,” “Nephropathy, Diabetic,” “Diabetic Glomerulosclerosis,” “Glomerulosclerosis, Diabetic.”

2.2. Literature screening criteria

Inclusion criteria: ① research subjects: subjects meet the diagnostic criteria of DN^[10]; ② intervention: the control group to control blood glucose, blood pressure, blood lipids, and other conventional integrated treatment, the experimental group in the control group on the basis of the intervention program with the addition of HKC treatment; ③ outcome indicators: primary outcome indicators: efficiency, blood creatinine (SCR), 24-hour urine protein concentration (24h-UTP), secondary outcome indicators: urea nitrogen (BUN), urinary albumin excretion rate (UAER), estimated glomerular filtration rate (eGFR), blood β2-microglobulin (β2-MG), urinary beta 2-microglobulin(Uβ2-MG), cystain C (Cys-C), fasting plasma glucose (FPG), 2-hour postprandial blood glucose (2hPG), glycated hemoglobin A1c (HbA1c), total triglyceride (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein-cholesterol (LDL-C), high-sensitivity C-reactive protein (hs-CRP), and interleukin-6 (IL-6). Safety index: incidence of adverse reactions. Study design: RCT. Exclusion criteria: ① animal experiments; ② repeated reports of the study; ③ the test group intervention in addition to the HKC also added other Chinese medicinal preparations; ④ the study subjects for other types of renal disease and dialysis patients with end-stage renal disease, combined with cardiovascular disease and other medication to affect the outcome indicators; ⑤ unpublished registered trial protocols, and the data of its research cannot be obtained; nonclinical RCT literature, such as case reports, summaries of experience, basic experiments, conference papers, retrospective studies, reviews; literature with obvious incomplete or incorrect information, or literature that cannot be combined, such as the absence of a clear course of treatment, the absence of endpoints, and so on.

2.3. Data extraction

EndNote X9 software was used to manage the retrieved literature. Two researchers independently screened the literature according to the inclusion and exclusion criteria, and any disagreement was referred to the corresponding author for adjudication. After deleting duplicates, the literature was first screened according to the title and abstract to exclude irrelevant literature, and then rescreened by reading the full text of the article to identify the studies to be included in the analysis. An Excel spreadsheet was set up to extract the data, which included: first author, year of publication, sample size, intervention, dose, duration of intervention, outcome indicators (mean and standard deviation), and reports of adverse events.

2.4. Literature quality assessment

Two researchers independently assessed the quality of the included studies according to the Cochrane Risk of Bias Assessment Tool,^[11] and disagreements were referred to the corresponding author for adjudication. Seven major biases were assessed: ① generation of randomized sequences; ② allocation concealment; ③ blinding of experimenters and subjects; ④ blinding of outcome assessors; ⑤ incomplete outcome data; ⑥ selective reporting; and ⑦ other biases. RevMan 5.3 software was used to map the quality of the literature.

2.5. Statistics and analysis of data

STATA/MP 17.0 and Review Manager 5.3 software were used for meta-analysis of the collected data. Dichotomous data were expressed by relative risk and 95% confidence interval (CI), and continuous data were expressed by standard mean difference (SMD) and 95% CI. Heterogeneity was determined by using I² and P-value, and when P > .1 or I² < 50%, a fixed-effects model was used, and if P < .1 or I² > 50%, a random-effects model was used. Subgroup analyses were performed according to clinical factors, Egger test were used to observe whether there was publication bias in each study.

3. Results

3.1. Basic characteristics of the included studies

A total of 1543 articles were retrieved from the literature, and 45 studies were included in the analysis according to the inclusion and exclusion criteria, and the detailed process of literature screening is shown in Figure 1. The published time interval of the included literature was 2010 to 2023, and all of them were in Chinese, with a total sample size of 4297 cases, among which 2168 cases were in the experimental group, and 2129 cases were in the control group, and the vast majority of them used 5 capsules tid, and the duration of treatment was from 4 weeks to 6 months. And the baseline data of all the included studies were comparable between groups, as shown in Table 1.

Figure 1.

PRISMA flow diagram of study inclusion and exclusion.

Table 1.

Basic characteristics of the included studies.

Study (author/year)	Sample size (T/C)	Baseline	Interventions T C		Dose	Duration of treatment	Main outcome
Xu (2016)[12]	62/62	Consistent	C + HKC	CT + Valsartan	5 capsules tid	6 months	②④⑤⑨⑩⑫⑬⑭⑮⑯
Hu (2021)[13]	46/46	Consistent	C + HKC	CT + Valsartan	5 capsules tid	56 days	①⑤⑩⑪⑬⑭
Li and Zhang (2014)[14]	42/42	Consistent	C + HKC	CT	5 capsules tid	56 days	②④⑩⑪⑫⑬⑭⑮⑯
Li (2020)[15]	58/54	Consistent	C + HKC	CT + Valsartan	2.5 g tid	6 months	②④⑤⑥⑧⑨⑩⑪⑫⑬⑭⑰⑱
Zou (2022)[16]	41/41	Consistent	C + HKC	CT + Telmisartan	5 capsules tid	3 months	①②③④⑨⑲
Guo (2022)[17]	57/57	Consistent	C + HKC	CT	5 capsules tid	56 days	①②③④⑦⑩⑫
Zhu (2020)[18]	40/40	Consistent	C + HKC	CTValsartan	5 capsules tid	3 months	①②④⑤⑲
Jin (2017)[19]	40/40	Consistent	C + HKC	CT	2.5 g tid	6 months	①②④⑤⑧⑬⑭⑯
Chen (2021)[20]	95/94	Consistent	C + HKC	CT + Acarbose	2.15 g tid	56 days	①②④⑩⑪⑫
Wang (2019)[21]	34/31	Consistent	C + HKC	CT + Valsartan	5 capsules tid	3 months	①②④⑤⑭⑲
Ke (2020)[22]	39/39	Consistent	C + HKC	CT + Valsartan	2.0 g tid	2 months	②④⑤
Wang (2017)[23]	60/60	Consistent	C + HKC	CT + Valsartan	5 capsules tid	2 months	①②⑤⑬⑭⑲
Li and Xie (2014)[24]	66/60	Consistent	C + HKC	CT	2.5 g tid	6 months	①②③④⑤⑧⑬⑭⑯
Xu et al (2018)[25]	50/50	Consistent	C + HKC	CT	5 capsules tid	6 months	②③④⑤⑧
Mao (2019)[26]	25/25	Consistent	C + HKC	CT + Calcium Dobesilate	2.5 g tid	3 months	②③④⑤⑲
Zhai (2020)[27]	64/63	Consistent	C + HKC	CT + Benazepril + irbesartan	2.5 g tid	56 days	①②④⑤⑧
Pu (2021)[28]	40/40	Consistent	C + HKC	CT + Telmisartan	2.5 g tid	84 days	①②③④⑨⑰⑱⑲
Zhao (2021)[29]	75/75	Consistent	C + HKC	CT + Valsartan	2.5 g tid	2 months	①②④⑤⑬⑭⑮⑯
Xu (2018)[30]	19/19	Consistent	C + HKC	CT + Benazepril	5 capsules tid	56 days	①②③④⑦⑨⑰⑱⑲
Qin (2016)[31]	48/48	Consistent	C + HKC	CT + Reduced glutathione	5 capsules tid	28 days	②③④⑧⑬⑭
Li (2022)[32]	48/48	Consistent	C + HKC	CT + Sitagliptin	5 capsules tid	2 months	②④⑲
Xu (2021)[33]	58/58	Consistent	C + HKC	CT + Simvastatin	5 capsules tid	84 days	②④⑤⑩⑫⑬⑭⑰⑱
Jiang (2023)[34]	44/44	Consistent	C + HKC	CT + Dapagliflozin	5 capsules tid	112 days	①②③④⑤⑩⑪⑫⑰⑱⑲
Jiang (2021)[35]	64/64	Consistent	C + HKC	CT	5 capsules tid	84 days	①②③④⑲
Feng (2016)[36]	60/60	Consistent	C + HKC	CT	5 capsules tid	56 days	①②③④⑩⑬⑰⑱
Sun (2018)[37]	64/64	Consistent	C + HKC	CT + Reduced glutathione	5 capsules tid	56 days	①②③⑤⑧⑩⑪⑫⑬⑭⑯
Chen (2015)[38]	30/30	Consistent	C + HKC	CT + Enalapril + irbesartan	5 capsules tid	3 months	②④⑤⑩⑫⑲
Hou (2018)[39]	40/40	Consistent	C + HKC	CT	5 capsules tid	3 months	①②③④⑩⑫⑬⑭
Hou (2021)[40]	43/43	Consistent	C + HKC	CT + Calcium Dobesilate	5 capsules tid	2 months	①②④⑰⑱
Yang (2020)[41]	32/32	Consistent	C + HKC	CT + Valsartan	2.5 g tid	2 months	①②④⑤
Liu (2018)[42]	50/42	Consistent	C + HKC	CT + Telmisartan	2.0 g tid	56 days	①②④⑤⑭⑰⑱
Li (2017)[43]	43/43	Consistent	C + HKC	CT	5 capsules tid	56 days	①⑤⑬⑭
Li (2014)[44]	48/47	Consistent	C + HKC	CT + Valsartan	5 capsules tid	42 days	②⑤⑬⑭⑲
Liang (2015)[45]	50/50	Consistent	C + HKC	CT + Valsartan	5 capsules tid	42 days	②⑤⑬⑭⑲
Liu (2022)[46]	73/73	Consistent	C + HKC	CT + Valsartan	3.5 g tid	56 days	①②③④⑤⑥⑧⑨⑰⑱⑲
Pan (2016)[47]	48/48	Consistent	C + HKC	CT	5 capsules tid	56 days	②③④
Li (2021)[48]	51/51	Consistent	C + HKC	CT + Valsartan	2.5 g tid	56 days	②④⑤⑩⑪⑫⑲
Zhang (2017)[49]	40/40	Consistent	C + HKC	CT + Valsartan	5 capsules tid	56 days	①②③
Huang (2016)[50]	48/48	Consistent	C + HKC	CT + Calcium Dobesilate	5 capsules tid	84 days	②④⑤⑬⑭⑲
Yang (2015)[51]	64/60	Consistent	C + HKC	CT	5 capsules tid	3 months	②③④⑧⑨⑫⑬⑭⑰
Jia (2015)[52]	38/32	Consistent	C + HKC	CT + Candesartan	2.0 g tid	28 days	②⑩⑫⑯⑲
Xue (2020)[53]	30/30	Consistent	C + HKC	CT + Alprostadil	2.5 g tid	56 days	①②④㉓
Du (2015)[54]	38/35	Consistent	C + HKC	CT	5 capsules tid	84 days	①②③⑩⑫⑬⑭⑲
Yang (2009)[55]	32/30	Consistent	C + HKC	CT	5 capsules tid	56 days	①②③④⑤⑦⑧
Teng (2021)[56]	31/31	Consistent	C + HKC	CT + Simvastatin	2.5 g tid	84 days	①②③④⑤⑥⑨⑩⑪⑫⑲

C = control group, CT = conventional therapy, T = experimental group.

① = clinically effective, ② = SCR, ③ = 24hUTP, ④ = BUN, ⑤ = UAER, ⑥ = GFR, ⑦ = β2-MG, ⑧ = Uβ2-MG, ⑨ = Cys-C, ⑩ = FBG, ⑪ = 2hPG, ⑫ = HbA1c, ⑬ = TG, ⑭ = TC, ⑮ = HDL-C, ⑯ = LDL-C, ⑰ = hs-CRP, ⑱ = IL-6, ⑲ = adverse reactions.

3.2. Risk of bias assessment of included studies

All included studies were randomized, with one study^[19] using the random envelope method of random assignment, one study^[21] using the red basketball method of random assignment, 25 studies^{[12–18,20,22–37,51]} using the table of random numbers method of random assignment, and 18 studies^{[38–50,52–56]} not explicitly indicating the method of randomization. All studies described at the end whether the allocation was hidden and whether it was double-talk, and all studies described at the end whether the outcome data were missing, and there was insufficient information to judge whether all data were selectively reported. The risk of bias assessment of the included studies is shown in Figures 2 and 3.

Figure 2.

Evaluation for bias risk of included studies.

Figure 3.

Risk of bias summary of included studies. “?” = unclear risk of bias, “–” = low risk of bias; “+” = high risk of bias.

3.3. Main outcome indicators

3.3.1. Efficient

Twenty-eight studies included reported clinical efficacy, with efficiency as the outcome index, but the assessment of efficacy varied, ranging from clinical symptoms and signs to renal function, urinary protein quantification, and urinary microalbumin excretion rate.^{[13,16–21,23,24,27–30,34–37,39–43,46,49,53–56]} Five of the studies^{[20,28,34,35,37]} used the Criteria for Diagnosis, Diagnostic Typing and Efficacy Assessment of Diabetic Nephropathy (Trial) as the criteria for assessing efficacy, while 3 studies^[19,24,30] used the Guidelines for Clinical Research of New Traditional Chinese Medicines as the criteria for assessment. Twenty studies^{[13,16–18,21,23,27,29,36,39–43,46,49,53–56]} were assessed by clinical signs and symptoms and laboratory indicators. Cumulatively, there were 1361 cases in the experimental group and 1337 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that the heterogeneity of the results of each study was small (P = .92, I² = 0%), so the analysis was performed using a fixed-effect model. The results showed that the addition of HKC had a significant efficacy in patients with DN [relative risk = 1.24, 95% CI (1.20, 1.29), P < .001], which was statistically significant, and the results were shown in Figure 4.

Figure 4.

Forest plot comparing the effects of HKC combined with conventional treatment versus conventional treatment alone.

3.3.2. SCR

Forty-three studies included^{[12,14–42,44–56]} The outcome indicators of SCR were reported with a cumulative total of 2079 cases in the experimental group and 2040 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was a statistical difference between the results of the studies (P < .001, I² = 94%), so the analysis was performed using a random-effects model. The results showed that the addition of HKC had a significant efficacy in reducing SCR [SMD = −1.19, 95% CI (−1.47, −0.91), P < .001]. According to the subgroups according to the treatment duration of the included population, the combined SMD value at 1 to 2 months of treatment was −0.76, 95% CI (−1.05, −0.48), P < .001, the combined SMD value at 3 months of treatment was −2.14, 95% CI (−2.94, −1.35), P < .001, and the combined SMD value at 4 to 6 months of treatment was −1.15, 95% CI (−1.64, −0.65), P < .001]. Subgroup analysis showed that heterogeneity still existed among subgroups, but the differences were statistically significant. The results are shown in Figure 5.

Figure 5.

Forest plot comparing the SCR of HKC combined with conventional treatment versus conventional treatment alone. SCR = serum creatinine.

3.3.3. 24-hour urine protein concentration

Twenty studies included^{[16,17,24–26,28,30,31,34–37,39,46,47,49,51,54–56]} reported outcome indicators of urinary protein quantification, with a cumulative total of 944 cases in the experimental group and 929 in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was a statistical difference between the results of the studies (P < .001, I² = 93%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a significant efficacy in lowering 24h-UTP [SMD = −1.27, 95% CI (−1.66, −0.88), P < .001]. According to the subgroups of the included population according to the duration of treatment, the combined SMD value at 1 to 2 months of treatment was −1.13, 95% CI (−1.86, −0.40), P = .002, and the combined SMD value at 3 months of treatment was −1.44, 95%CI (−1,87, −1,01), P < .001, and the combined SMD value at 4 to 6 months of treatment was −1.22, 95% CI (−2.04, −0.40), P = .003). Subgroup analysis showed that heterogeneity still existed among subgroups, but the differences were statistically significant. The results are shown in Figure 6.

Figure 6.

Forest plot comparing the 24h-UTP of HKC combined with conventional treatment versus conventional treatment alone. 24h-UTP = 24-hour urine protein concentration.

3.4. Secondary outcome indicators

3.4.1. Blood urea nitrogen

Thirty-six included studies^{[12,14–22,24–36,38–42,46–48,50,51,53,55,56]} reported on the outcome metrics of BUN. The cumulative number of cases in the experimental group was 1741, and the control group was 1712. Based on the heterogeneity test of Review Manager 5.3, the results of the studies were statistically different from each other (P < .001, I² = 89%), and were analyzed using a random-effects model: the results showed that the addition of HKC had a significant efficacy in lowering the BUN [SMD = −0.79, 95% CI (−1.00, −0.57), P < .001]. The subgroups were divided according to the duration of treatment of the included population. The combined SMD value at 1 to 2 months of treatment was −0.64, 95% CI (−0.86, −0.43), P < .001, the combined SMD value at 3 months of treatment was −1.02, 95% CI (−1.52, −0.53), P < .001, and the combined SMD value at 4 to 6 months of treatment was −0.81, 95%CI (−1.48, −0.13), P = .02. Subgroup analysis showed that heterogeneity still existed among subgroups. But the differences were statistically significant. The results are shown in Figure 7.

Figure 7.

Forest plot comparing the BUN of HKC combined with conventional treatment versus conventional treatment alone. BUN = blood urea nitrogen.

3.4.2. Urinary albumin excretion rate

Twenty-seven included studies^{[12,13,15,18,19,21–27,29,33,34,37,38,41–46,48,50,55,56]} reported outcome indicators of UAER, with a cumulative total of 1313 cases in the experimental group and 1288 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P < .001, I² = 96%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a clear efficacy in reducing UAER [SMD = −1.90, 95% CI (−2.35, −1.44), P < .001]. The subgroups were divided according to the duration of treatment of the included population, and the combined SMD value at 1 to 2 months of treatment was −1.23, 95% CI (−1.78, −0.68), P < .001, and the combined SMD value at 3 months of treatment was −2.12, 95% CI (−2.85, −1.39), P < .001, and the combined SMD value at 4 to 6 months of treatment was −3.31, 95% CI (−4.73, −1.89), P < .001. And the subgroup analysis showed that statistical significance existed in each subgroup. The results are shown in Figure 8.

Figure 8.

Forest plot comparing the UAER of HKC combined with conventional treatment versus conventional treatment alone. UAER = urinary albumin excretion rate.

3.4.3. Estimated glomerular filtration rate

Three included studies^[15,46,56] reported outcome indicators of eGFR, with a cumulative total of 162 cases in the experimental group and 158 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P = .001, I² = 85%), so the analysis was performed using a random effects model. The results showed no difference between the experimental and control groups in improving eGFR [SMD = −0.21, 95% CI (−0.80, 0.39), P = .50], as shown in Figure 9. Subgroup analyses could not be developed due to limitations in the number of studies on this metric.

Figure 9.

Forest plot comparing the eGFR of HKC combined with conventional treatment versus conventional treatment alone. eGFR = estimated glomerular filtration rate.

3.4.4. Beta 2-microglobulin

Three included studies^[17,30,55] reported outcome indicators of β2-MG, with a cumulative total of 108 cases in the experimental group and 106 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P = .22, I² = 33%), so the analysis was performed using a fixed effects model. The results showed that the addition of HKC had a clear efficacy in reducing β2-MG [SMD = −0.29, 95% CI (−0.56, −0.02), P = .03]. As shown in Figure 10. Subgroup analyses could not be developed due to limitations in the number of studies on this metric.

Figure 10.

Forest plot comparing the β2-MG of HKC combined with conventional treatment versus conventional treatment alone. β2-MG = beta 2-microglobulin.

3.4.5. Urinary beta 2-microglobulin

Ten included studies^{[15,19,24,25,27,31,37,46,51,55]} reported outcome indicators of Uβ2-MG, with a cumulative total of 559 cases in the experimental group and 542 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P < .001, I² = 91%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a clear efficacy in reducing Uβ2-MG [SMD = −1.04, 95% CI (−1.46, −0.62), P < .001]. The subgroups were divided according to the duration of treatment of the included population, and the combined SMD value at 1 to 2 months of treatment was −0.61, 95% CI (−0.93, −0.30), P < .001, and the combined SMD value at 3 months of treatment was −0.81, 95% CI (−1.17, −0.44), P < .001, and the combined SMD value at 4 to 6 months of treatment was −1.67, 95% CI (−2.60, −0.74), P < .001. And the subgroup analysis showed that statistical significance existed in each subgroup. The results are shown in Figure 11.

Figure 11.

Forest plot comparing the Uβ2-MG of HKC combined with conventional treatment versus conventional treatment alone. Uβ2-MG = urinary beta 2-microglobulin.

3.4.6. Cystatin C

Eight included studies^{[12,15,16,28,30,46,51,56]} reported outcome indicators of Cys-C, with a cumulative total of 388 cases in the experimental group and 380 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P < .001, I² = 89%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a clear efficacy in reducing Cys-C [SMD = −1.20, 95% CI (−1.69, −0.71), P < .001]. The subgroups were divided according to the duration of treatment of the included population, and the combined SMD value at 1 to 2 months of treatment was −0.85, 95% CI (−1.28, −0.42), P < .001, and the combined SMD value at 3 months of treatment was −1.20, 95% CI (−1.97, −0.43), P = .002, and the combined SMD value at 4 to 6 months of treatment was −1.61, 95% CI (−3,04, −0.17), P < .001. The results are shown in Figure 12.

Figure 12.

Forest plot comparing the Cys-C of HKC combined with conventional treatment versus conventional treatment alone. Cys-C = cystatin C.

3.4.7. Fasting blood glucose (FBG)

Sixteen included studies^{[12–15,17,20,33,34,36–39,48,52,54,56]} reported outcome indicators of FBG, with a cumulative total of 814 cases in the experimental group and 800 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P < .001, I² = 88%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a clear efficacy in reducing FBG [SMD = −0.85, 95% CI (−1.15, −0.55), P < .001]. The subgroups were divided according to the duration of treatment of the included population, and the combined SMD value at 1 to 2 months of treatment was −0.82, 95% CI (−1.30, −0.34), P < .001, and the combined SMD value at 3 months of treatment was −1.04, 95% CI (−1.62, −0.47), P < .001, and the combined SMD value at 4 to 6 months of treatment was −0.65, 95% CI (−1.01, −0.30), P < .001. And the subgroup analysis showed that statistical significance existed in each subgroup. The results are shown in Figure 13.

Figure 13.

Forest plot comparing the FBG of HKC combined with conventional treatment versus conventional treatment alone. FBG = fasting blood glucose.

3.4.8. 2-hour postprandial blood glucose

Eight included studies^{[13–15,20,34,37,48,56]} reported outcome indicators of 2hPG, with a cumulative total of 431 cases in the experimental group and 426 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P < .001, I² = 82%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a clear efficacy in reducing 2hPG [SMD = −0.80, 95% CI (−1.14, −0.46), P < .001]. The subgroups were divided according to the duration of treatment of the included population, and the combined SMD value at 1 to 2 months of treatment was −0.90, 95% CI (−1.43, −0.38), P < .001, and the combined SMD value at 3 months of treatment was −0.93, 95% CI (−1.46, −0.41), P < .001, and the combined SMD value at 4 to 6 months of treatment was −0.50, 95%CI (−0.78, −0.22), P < .001. And the subgroup analysis showed that statistical significance existed in each subgroup. The results are shown in Figure 14.

Figure 14.

Forest plot comparing the 2hPG of HKC combined with conventional treatment versus conventional treatment alone. 2hPG = 2-hour postprandial blood glucose.

3.4.9. Hemoglobin A1C

Fifteen included studies^{[12,14,15,17,20,33,34,37–39,48,51,52,54,56]} reported outcome indicators of HbA1c, with a cumulative total of 772 cases in the experimental group and 754 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P < .001, I² = 81%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a clear efficacy in reducing HbA1c [SMD = −0.76, 95% CI (−1.00, −0.51), P < .001]. The subgroups were divided according to the duration of treatment of the included population, and the combined SMD value at 1 to 2 months of treatment was −0.62, 95% CI (−0.95, −0.29), P < .001, and the combined SMD value at 3 months of treatment was −0.84, 95% CI (−1.19, −0.49), P < .001, and the combined SMD value at 4 to 6 months of treatment was −0.86, 95% CI (−1.78, −0.06), P < .001. And the subgroup analysis showed that statistical significance existed in each subgroup. The results are shown in Figure 15.

Figure 15.

Forest plot comparing the HbA1c of HKC combined with conventional treatment versus conventional treatment alone. HbA1c = hemoglobin A1C.

3.4.10. Triglyceride

Nineteen included studies^{[12–15,19,23,24,29,31,33,36,37,39,43–45,50,51,54]} reported outcome indicators of TG, with a cumulative total of 1010 cases in the experimental group and 992 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P < .001, I² = 87%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a clear efficacy in reducing TG [SMD = −0.81, 95% CI (−1.07, −0.55), P < .001]. The subgroups were divided according to the duration of treatment of the included population, and the combined SMD value at 1 to 2 months of treatment was −0.97, 95% CI (−1.41, −0.54), P < .001, and the combined SMD value at 3 months of treatment was −0.58, 95% CI (−1.06, −0.11), P < .001, and the combined SMD value at 4-6 months of treatment was −0.72, 95% CI(−0.94, −0.50), P < .001. And the subgroup analysis showed that statistical significance existed in each subgroup. The results are shown in Figure 16.

Figure 16.

Forest plot comparing the TG of HKC combined with conventional treatment versus conventional treatment alone. TG = triglyceride.

3.4.11. Total cholesterol

Nineteen included studies^{[12–15,19,23,24,29,31,33,37,39,42–45,50,51,54]} reported outcome indicators of TC, with a cumulative total of 1000 cases in the experimental group and 974 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P < .001, I² = 95%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a clear efficacy in reducing TC [SMD = −1.24, 95% CI (−1.68, −0.81), P < .001]. The subgroups were divided according to the duration of treatment of the included population, and the combined SMD value at 1 to 2 months of treatment was −1.77, 95% CI (−2.55, -0.98), P < .001, and the combined SMD value at 3 months of treatment was −0.51, 95% CI (−0.90, −0.13), P = .008, and the combined SMD value at 4 to 6 months of treatment was −0.93, 95% CI (−1.58, −0.29), P = .005. and the subgroup analysis showed that statistical significance existed in each subgroup. The results are shown in Figure 17.

Figure 17.

Forest plot comparing the TC of HKC combined with conventional treatment versus conventional treatment alone. TC = total cholesterol.

3.4.12. High-density lipoprotein

Three included studies^[12,14,29] reported outcome indicators of HDL-C, with a cumulative total of 179 cases in the experimental group and 179 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P < .001, I² = 91%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a clear efficacy in reducing HDL-C [SMD = 1.08, 95% CI (0.30, 1.85), P = .006], as shown in Figure 18. Subgroup analyses could not be developed due to limitations in the number of studies on this metric.

Figure 18.

Forest plot comparing the HDL-C of HKC combined with conventional treatment versus conventional treatment alone. HDL-C = high-density lipoprotein.

3.4.13. Low density lipoprotein-cholesterol

Seven included studies^{[12,14,19,24,29,37,52]} reported outcome indicators of LDL-C, with a cumulative total of 387 cases in the experimental group and 375 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P < .001, I² = 93%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a clear efficacy in reducing LDL-C [SMD = −1.12, 95% CI (−1.71, −0.53), P < .001]. The subgroups were divided according to the duration of treatment of the included population, and the combined SMD value at 1 to 2 months of treatment was −1.41, 95% CI (−2.43, −0.39), P = .007, and the combined SMD value at 4 to 6 months of treatment was −0.75, 95% CI (−1.16, −0.34), P < .001. And the subgroup analysis showed that statistical significance existed in each subgroup. The results are shown in Figure 19.

Figure 19.

Forest plot comparing the LDL-C of HKC combined with conventional treatment versus conventional treatment alone. LDL-C = low density lipoprotein-cholesterol.

3.4.14. High-sensitivity C-reactive protein

Eight included studies^{[15,28,34,36,40,42,46,51]} reported outcome indicators of hs-CRP, with a cumulative total of 432 cases in the experimental group and 416 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P < .001, I² = 95%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a clear efficacy in reducing hs-CRP [SMD = −1.51, 95% CI (−2.24, −0.77), P < .001]. The subgroups were divided according to the duration of treatment of the included population, and the combined SMD value at 1 to 2 months of treatment was −1.87, 95% CI (−2.69, −1.05), P < .001, and the combined SMD value at 3 months of treatment was −1.71, 95% CI (−4.59, −1.17), P = .24, and the combined SMD value at 4 to 6 months of treatment was −0.61, 95% CI (−1.33, 0.12), P = .10. And the subgroup analysis showed that statistical significance existed in each subgroup. There was no statistical significance in the use of capsules in the subgroup of 3, 4 to 6 months of treatment compared with the control group, while there was statistical significance in the remaining subgroups. The results are shown in Figure 20.

Figure 20.

Forest plot comparing the hs-CRP of HKC combined with conventional treatment versus conventional treatment alone. hs-CRP = high-sensitivity C-reactive protein.

3.4.15. Interleukin-6

Nine included studies^{[15,28,30,33,34,36,40,42,46]} reported outcome indicators of IL-6, with a cumulative total of 445 cases in the experimental group and 433 cases in the control group. The heterogeneity test based on Review Manager 5.3 suggested that there was statistical heterogeneity between the results of the studies (P < .001, I² = 95%), so the analysis was performed using a random effects model. The results showed that the addition of HKC had a clear efficacy in reducing IL-6 [SMD = −2.30, 95% CI (−3.07, −1.53), P < .001]. The subgroups were divided according to the duration of treatment of the included population, and the combined SMD value at 1 to 2 months of treatment was −3.38, 95% CI (−4.95, −1.81), P < .001, and the combined SMD value at 3 months of treatment was −1.56, 95% CI (−1.88, −1.23), P < .001, and the combined SMD value at 4 to 6 months of treatment was −1.81, 95% CI (−2.14, −1.48), P < .001. And the subgroup analysis showed that statistical significance existed in each subgroup. The results are shown in Figure 21.

Figure 21.

Forest plot comparing the IL-6 of HKC combined with conventional treatment versus conventional treatment alone. IL-6 = interleukin-6.

3.5. Analysis of adverse reactions

Twenty-one included studies^{[16,18,21,23,26,28,30–32,34,35,38,44–46,48,50,52–54,56]} reported adverse reactions, with a cumulative total of 900 cases in the experimental group, and control group 887 cases. The heterogeneity test based on Review Manager 5.3 suggested that there was little heterogeneity in the results of the studies (P = .85, I² = 0%), so the analysis was purchased with a fixed effect model. The results showed that there was no statistically significant difference in the incidence of adverse reactions between the study group with the addition of HKC and the control group, and the results are shown in Figure 22.

Figure 22.

Forest plot comparing the adverse reactions of HKC combined with conventional treatment versus conventional treatment alone.

3.6. Publication bias

Publication bias analysis was performed on the efficiency, Egger test for publication bias t = 7.39, P = .000, suggesting that there was a bias, but the specific reason for the bias could not be determined. Publication bias analysis was performed on SCR, Egger test for publication bias t = −1.09, P = .282, suggesting that there was no publication bias. In the publication bias analysis of 24h-UTP, Egger test for publication bias t = -3.24, P = .005, indicating publication bias, suggesting a lack of studies with negative results.

4. Discussions

HKC is a new Chinese medicine developed based on the theory of Chinese medicine and the application of modern research technology, its function is to clear heat and relieve dampness, detoxification and detumescence. The main components are flavonoids, modern pharmacological research found that the drug has: anti-inflammatory, diuretic, vasodilating, relieving vasospasm, reducing platelet aggregation Modern pharmacological research has found that the drug has the following effects: anti-inflammatory, diuretic, vasodilator, relieving vasospasm, antiplatelet aggregation, reducing glomerular immune-inflammatory reaction, promoting immune complex clearance, protecting renal tubular function and so on. Therefore, in recent years, HKC has been widely used in the treatment of various renal diseases, including the treatment of DN. In this study, the efficacy and safety of HKC in the treatment of DN were systematically evaluated by meta-analysis of previous RCT studies. A total of 45 RCT studies were included in this study, of which 21 were added with ACEI/ARB,^{[12,13,15,16,18,21–23,27–30,38,41,42,44–46,48,49,52,53]} one each for the addition of acarbose, selegiline, simvastatin, dagliflozin, and prostaglandin.^[20,32–34] There were 2 additions of reduced glutathione,^[31,37] 3 items with the addition of Calcium Dobesilate,^[26,40,50] the rest are western medicine based treatments.

Based on the positive results of HKC in improving proteinuria in patients with chronic glomerulonephritis, this study focused on the effect of HKC on the outcome of proteinuria in DN, the experimental group compared to the control group in improving the quantitative urine protein, UAER with significant efficacy compared to the control group. Subgroup analyses revealed that the improvement in urine protein quantification was more pronounced in the 3 months course than in the 1 to 2 months course, while there was no significant difference in UAER. This finding is consistent with a 2020 multi-RCT study of Abelmoschus manihot in the treatment of IgA Nephropathy.^[57] This study found that the most significant improvement in urinary protein quantification was observed at 0 to 3 months of use of Abelmoschus manihot, with a slow and effective reduction in urinary protein quantification occurring after 3 months. The RCTs included in this study had small sample sizes and high heterogeneity, and the effect of treatment duration on urine protein quantification still needs to be further explored in high-quality studies.

In this study, we also observed the outcome indicators of renal function, and the outcome indicators of SCR and BUN showed that the efficacy of the experimental group was better than that of the control group, but there was no difference between the experimental group and the control group in terms of the increase of eGFR, which may be due to the small sample size, and needs to be further confirmed. In this study, we observed the effects on sensitive indexes such as β2-MG, Uβ2-MG, Cys-C, etc, and found that the ending indexes of the experimental group were better than those of the control group, suggests that the addition of HKC exerted a favorable improvement on these indicators. In this study, we also investigated the related indexes of glucose metabolism and lipid metabolism, and the results of the study showed that the experimental group performed better than the control group both in terms of FBG, 2hPG, HbA1c, and TG, TC, HDL-C, and LDL-C. The pathogenesis of DN is complex, and an increasing number of studies have shown that inflammation plays an important role in the development of DN. A renal pathology study on DN^[58] found that immune cell infiltration, mainly macrophages, was prevalent in renal tissues at all stages of DN. Also, study^[59] found that inflammatory factor IL-6, tumor necrosis factor, etc were also observed in renal biopsy specimens of DN patients. Based on the above studies, the present study observed the outcome indicators such as hs-CRP and IL-6, and the results showed that HKC was superior to the control group in regulating inflammatory factors, and HKC might delay the progression of DN by regulating inflammatory factors. In terms of safety, there was no statistically significant difference between the addition of HKC group and the control group, and the included adverse reactions were mainly gastrointestinal symptoms and malaise, and no serious adverse reactions were seen. However, the RCTs included in this study had a relatively short observation period, all within half a year, and the safety of its long-term use could not be evaluated for the time being.

A 2015 meta-analysis of HKC in the treatment of DN^[60] found that HKC combined with ACEI or ARB drugs had the efficacy of better regulation of 24h-UTP, UAER, SCR, BUN, TC, TG, etc than ACEI or ARB drugs alone, and consistent results were obtained in the present study, but on FBG, different conclusions were drawn, which may be the reason why the original article included fewer papers than this paper. Meanwhile, the present study found that the HKC also has good efficacy on the regulation of inflammatory factors. Regarding urinary protein quantification, this study found that the addition of HKC was superior to the control group, but the effect of the course of treatment on urinary protein quantification is still uncertain, and we look forward to more high-quality studies with a long follow-up period.

5. Limitations of the current study

There are still many shortcomings in this study, mainly including: (1) most of the included studies did not provide specific descriptions of the risks in the literature, such as hidden randomization, use of blinding, missing data, and publication bias, etc, (2) the inclusion of studies with small sample sizes and publication bias made the results of this study less stable, (3) the follow-up time of the included studies was generally less than half a year, and it was not possible to evaluate the long-term safety index of HKC for the treatment of DN, and there was a lack of reliable endpoints, (4) given that all included literature were in Chinese and the trials were completed in Chinese units, the results have some limitations.

6. Conclusions

Therefore, although the available evidence is not yet sufficient to draw convincing conclusions, the combined use of HKC in the treatment of DN improved patients’ conditions to a certain extent, but in order to better evaluate the efficacy and safety of HKC as an adjunctive intervention in DN, more rigorously designed large-sample, multicenter RCTs are needed in the future to provide evidentiary support, as well as to standardize the reporting of adverse reactions and to emphasize the negative results, in order to clarify its efficacy.

7. Ethics and dissemination

Patients and the public were not required to participate in the research design, data statistical process and results of this study. Informed consent and ethical approval were not required. In addition, the results of this study will be disseminated through peer-reviewed publications.

Author contributions

Conceptualization: Yuandong Li.

Data curation: Cui Wu, Chaoqun Song.

Investigation: Yuandong Li.

Methodology: Cui Wu.

Project administration: Zheng Nan.

Resources: Chaoqun Song.

Supervision: Shilin Liu.

Writing – original draft: Yuandong Li.

Writing – review & editing: Yuandong Li.